Fecal incontinence (FI) is a socially devastating symptom in older women and may contribute to institutionalization. The epidemiology and pathophysiology of "idiopathic" FI is incompletely understood. Current concepts based on tertiary-care studies heavily emphasize the role of anal sphincter defects visualized by endoanal ultrasound. Preliminary studies suggest that the prevalence of FI in Olmsted County in women greater than or equal to 50 years is 17.8 percent with a median age of onset of 61 years. Obstetric events, diarrhea/urgency and obesity are risk factors for FI. Our novel "fluoroscopic" single-shot fast spin-echo MRI techniques visualize pelvic floor descent during defecation in real-time. In contrast to US, endoanal MRI depicts external sphincter defects and atrophy, puborectalis thinning and global pelvic floor laxity. The hypothesis is that fecal incontinence is not attributable to obstetric trauma alone, but the cumulative result of pelvic floor weakness caused by obstetric trauma, excessive straining, obesity, aging and menopause, compounded by diarrhea. This hypothesis will be addressed by combining the data infrastructure of the Rochester Epidemiology Project with state- of-the-art physiological measurements in a community-based sample. A questionnaire will be mailed to a cohort of approximately 1,000 women surveyed previously to ascertain the incidence and natural history of FI, and, a new sample of 5,000 women to determine the prevalence and frequency of FI. Putative risk factors for pelvic floor injury (obstetric trauma, chronic straining, obesity and estrogen depletion) and FI (diarrhea and rectal urgency) will be evaluated in a case-control study of approximately 200 patients with FI at least once a month and approximately 200 controls. approximately 100 patients with FI and approximately 100 controls will have MRI fluoroscopy to characterize the specific global pelvic floor abnormality (i.e., anal sphincter defects, sphincter atrophy, puborectalis thinning and pelvic floor laxity) associated with FI. These studies will refine our understanding of the epidemiology of FI, identify the obstetric risk factors responsible for delayed manifestations of pelvic floor injury, i.e. FI, underscore the importance of non- obstetric risk factors for FI and provide novel insights into the specific pattern of pelvic floor injury associated with FI in a community. These steps are necessary for reducing the incidence of pelvic floor damage by risk factor modification, identifying patients at higher risk of progressing to symptomatic FI, and designing appropriate interventions to halt this process.